CN107728469B - A kind of laser particle control method - Google Patents
A kind of laser particle control method Download PDFInfo
- Publication number
- CN107728469B CN107728469B CN201710752460.3A CN201710752460A CN107728469B CN 107728469 B CN107728469 B CN 107728469B CN 201710752460 A CN201710752460 A CN 201710752460A CN 107728469 B CN107728469 B CN 107728469B
- Authority
- CN
- China
- Prior art keywords
- laser particle
- particle control
- monomer
- control monomer
- reverse learning
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 239000002245 particle Substances 0.000 title claims abstract description 187
- 238000000034 method Methods 0.000 title claims abstract description 36
- 239000000178 monomer Substances 0.000 claims abstract description 164
- 230000000739 chaotic effect Effects 0.000 claims abstract description 6
- 239000013598 vector Substances 0.000 claims description 20
- 238000004364 calculation method Methods 0.000 claims description 4
- 239000002131 composite material Substances 0.000 claims description 4
- 238000009616 inductively coupled plasma Methods 0.000 claims description 3
- 230000009467 reduction Effects 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 230000005855 radiation Effects 0.000 claims description 2
- 238000004422 calculation algorithm Methods 0.000 description 16
- 238000005381 potential energy Methods 0.000 description 9
- 238000013461 design Methods 0.000 description 8
- 238000007906 compression Methods 0.000 description 7
- 230000006399 behavior Effects 0.000 description 4
- 241000283153 Cetacea Species 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 3
- 230000005284 excitation Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000029264 phototaxis Effects 0.000 description 3
- 238000010521 absorption reaction Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 2
- 230000033001 locomotion Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 241000544061 Cuculus canorus Species 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Data Mining & Analysis (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Automation & Control Theory (AREA)
- Lasers (AREA)
Abstract
The invention discloses a kind of laser particle control methods, are initialized to monomer coordinate in cluster using reversion choice；It is re-introduced into IRICP and seeks the optimal laser particle control monomer of arrow method searching；Meanwhile Laplce's CHAOTIC INTERFERENCE is carried out to optimal laser particle control monomer in current cluster, to obtain the optimal monomer of optimal laser particle in cluster.The present invention can be compared to each other corresponding laser particle in multiple clusters when fitness compares by wheel, can also be compared to internal laser particle in a cluster, reach the requirement of control accuracy.
Description
Technical Field
The invention relates to the field of computers, in particular to a laser particle control method.
Background
The cluster intelligent control algorithm is a heuristic control method for simulating a certain evolution mechanism or a certain activity behavior in the nature, and does not need to rely on the gradient information of the problem, so that the method for solving the control problem by utilizing the cluster intelligent control algorithm becomes a research hotspot in the field of evolutionary computation. Aiming at the multi-target control problem, the tangkuzon and the like provide an algorithm for ensuring cluster diversity by introducing a near-Neighbor Function Criterion (NFC). Mirjalili and Lewis simulate the predation behavior of whales with whales, and provide a whale control algorithm. The application and development of the method provide a reinforced multi-target quantum behavior particle swarm control algorithm. Li et al propose an efficient cuckoo vector-finding algorithm based on an orthogonal learning method.
Yang of Cambridge university simulates the phototactic behavior of laser particle control in nature, and proposes a laser particle control Algorithm (LP). LP is a meta-heuristic control algorithm derived by simulating phototaxis behavior controlled by laser particles in nature, and utilizes phototaxis control characteristics of the laser particles to search laser particle control with better coordinates (stronger photon stimulated intensity) in a vector searching space and approach the laser particle control to the vector searching space, thereby achieving the purpose of controlling and searching vectors. However, when the vector is searched in space, only some laser particle points with larger errors can be searched, and the requirement of accuracy cannot be met.
Disclosure of Invention
In view of the above-mentioned drawbacks of the prior art, the present invention provides a laser particle control method, which greatly improves the optimization while maintaining the diversity of the clusters.
The invention provides a laser particle control method, which is improved in that a reverse learning method is adopted to initialize a monomer coordinate in a cluster; then an IRICP vector searching method is introduced to search for an optimal laser particle control monomer group; meanwhile, Laplace chaotic interference is carried out on the optimal laser particle control monomer group in the current cluster, so that the optimal monomer of the optimal laser particle control monomer group in the cluster is obtained.
Preferably, the step of initializing coordinates of the single body in the cluster by using inverse learning includes:
setting the population scale in the cluster as N, wherein any laser particle control monomer is as follows:
x∈[l,u]；
its inverse solution x' is:
x′＝l+u-x；
if any candidate laser particle control monomer in the D-dimension vector searching space is P, then:
P＝(x1,x2,…,xd)；
wherein xi∈[ai,bi]I is 1,2, …, d; then its corresponding inverse solution P' is:
P′＝(x1′,x2′,…,x′d)；
wherein xi′＝li+ui-xi；
And performing reverse learning initialization to obtain the corresponding fitness in the cluster, sequencing the fitness values, and selecting the first N laser particle control monomers corresponding to the fitness values as the initial cluster.
Preferably, the step of finding the optimal laser particle control monomer group by introducing an IRICP vector searching method comprises:
(1) setting:
initial laser particle control monomer x before reverse learning initialization1，x1∈Rn；
Unit orthogonal direction dnTaking a coordinate direction;
step size
the expansion factor alpha, alpha is more than 1;
reduction factor beta, β ∈ (-1, 0);
the allowable error epsilon is more than 0; order:
i is 1 to n;
in the formula, y1Controlling a monomer for the initialized initial laser particles after reverse learning; k is a serial number and takes the value of 1 to n; i and j are laser particle control monomer variables, j is 1 to n, and i is 1 to n;
(2) laser particle control monomer y after reverse learning initialization with interference in actual judgmentjWhether the corresponding fitness value is smaller than the fitness value corresponding to the initial laser particle control monomer after the reverse learning initialization, namely f (y)j+δjdj)＜f(yj) If yes, the laser particle control monomer y after the reverse learning initialization is enabledjThe next laser particle control monomer is equal to the current laser particle control monomer, and the laser particle control monomer yjis enlarged by a factor of a, i.e. yj+1＝yj，δj＝αδj(ii) a Otherwise, the laser particle control monomer y after the reverse learning initialization is enabledjThe next laser particle control monomer is equal to the current laser particle control monomer, and the laser particle control monomer yjby a factor of beta, i.e. yj+1＝yj，δj＝βδj；
(3) Judging a laser particle control monomer variable j, if the laser particle control monomer variable j is smaller than n, namely j is smaller than n, enabling the next laser particle control monomer variable of the laser particle control monomer variable j to be equal to the current laser particle control monomer variable, namely j equals to j +1, and returning to the step (2); otherwise, entering the step (4);
(4) judging whether the fitness value corresponding to the laser particle control single body of the n +1 point after the reverse learning initialization is smaller than the fitness value corresponding to the initial laser particle control single body after the reverse learning initialization, namely f (y)n+1)＜f(y1) If yes, the initial laser particle control monomer after the reverse learning initialization is equal to the laser particle control monomer at the n +1 point after the reverse learning initialization, namely y1＝yn+1J is 1, and return to step (2); if the fitness value corresponding to the laser particle control monomer of the n +1 point after the reverse learning initialization is equal to the fitness value corresponding to the initial laser particle control monomer after the reverse learning initializationOf (a), i.e. f (y)n +1)＝f(y1) Entering the step (5);
(5) judging whether the fitness value corresponding to the laser particle control monomer of the n +1 point after the reverse learning initialization is smaller than the fitness value corresponding to the laser particle control monomer of the k point after the reverse learning initialization, namely f (y)n+1)＜f(yk) Entering the step (6); otherwise, judging all laser particle control monomer variables j, and if the step length of the point of the laser particle control monomer variable j is not larger than the allowable error, namely | delta | (delta [)jIf | ≦ epsilon, ending, and taking the laser particle control monomer at the k point before the reverse learning initialization as the optimal monomer, otherwise, taking the initial laser particle control monomer after the reverse learning initialization equal to the laser particle control monomer at the n +1 point after the reverse learning initialization, namely y1＝yn+1And making the laser particle control monomer variable j equal to 1, and returning to the step (2);
(6) the initial laser particle control monomer of k +1 point before the reverse learning initialization is equal to the initial laser particle control monomer of n +1 point after the reverse learning initialization, namely xk+1＝yn+1Judging whether the corresponding fitness value of the initial laser particle control single body of the k +1 point before the reverse learning initialization and the initial laser particle control single body of the k point before the reverse learning initialization is not more than an allowable error, namely | xk+1-xkIf | | is less than or equal to epsilon, taking the initial laser particle control monomer of the k +1 point before the reverse learning initialization as a minimum point and a control object, and finishing the calculation; otherwise, entering the step (7);
(7) order:
in the formula, λiAll along the unit orthogonal direction diAlgebraic sum of step sizes of (a);
defining a set of directions p1,p2,...,pn：
Using Gram-Schmidt orthogonalization method to set the direction { p }jOrthogonalization, let:
then the direction group { pjUnitization, let:
controlling unit orthogonal direction average of a single j point for laser particles; obtaining n new orthogonal vector searching directions according to the formula;
(8) order:
wherein j is 1 to n; and j is 1, then:
y1＝xk+1；
k＝k+1；
and (4) returning to the step (2).
Preferably, the step of performing laplacian chaos interference on the optimal laser particle control monomer in the current cluster includes:
control monomer for optimal laser particles at time t in current clusterThe laplace interference strategy is implemented as follows:
in the formula,representing the coordinates of the laser particle control monomer after the interference, wherein Gaussion (sigma) is a random variable satisfying Laplace distribution; the composite optimal coordinate is updated as follows:
in the formula,controlling the fitness value corresponding to the coordinates of the monomer for the interfered laser particles;controlling monomer for optimal laser particles at time t in current clusterA corresponding fitness value;
by controlling the monomer for the current optimum laser particleAnd carrying out interference operation, thereby obtaining the optimal monomer of the optimal laser particle control monomer group in the cluster.
In the technical scheme of the invention, when the fitness is compared by turns, the corresponding laser particles in a plurality of clusters can be compared with each other, and the internal laser particles can be compared in one cluster, so that the requirement on control precision is met.
Drawings
FIG. 1 is a flow chart of an embodiment of the present invention;
FIG. 2 is a schematic structural diagram illustrating a design problem of a tension-compression nanometer elastic potential energy machine according to an embodiment of the present invention;
FIG. 3 is an optimal monomer comparison data obtained by the eight-cluster intelligent method according to the embodiment of the present invention;
fig. 4 is a comparison of statistical results obtained by the eight-cluster intelligent method according to the embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail below with reference to the accompanying drawings by way of examples of preferred embodiments. It should be noted, however, that the numerous details set forth in the description are merely for the purpose of providing the reader with a thorough understanding of one or more aspects of the present invention, which may be practiced without these specific details.
the traditional laser particle control algorithm is generally based on the following assumptions that (1) the guided photon with strong photon excitation intensity is weak, (2) the guidance degree β controlled by the laser particle is in direct proportion to the photon excitation intensity I, and the β and the I become smaller along with the increase of the distance r, and (3) the photon excitation intensity is related to an objective function.
Determining laser particle control monomer xiWith laser particle controlling monomer xjDistance r betweenij：
Where d is the dimension of the decision variable. XiTo control monomer xiThe position of (a); xjTo control monomer xjThe position of (a); x is the number ofikControl unit x with serial number k from i to ni；xjkControl unit x with serial number k from j to niOf which is in contact with xikthe guidance degree β of the laser particle control is as follows:
wherein, beta0∈[0,1]Is rij0, γ ∈ [0,10 ]]Is the laser absorption coefficient;
the coordinates of the laser particle control are moved and updated. The laser particle control monomer i is guided by the laser particle control monomer j with brighter photon excited intensity to generate coordinate movement:
wherein,represents the coordinate of the ith laser particle control monomer at the t moment, and belongs to [0,1 ]]Random step length, rand-U (0,1) is random number;
the laser particle with the brightest photon excited intensity controls random motion:
in the formula,is the most in the cluster at the t-th timeThe monomer coordinates are optimized, and the rand is from 0 to 1 for loop iteration.
In the laser particle control method provided by this embodiment, a flowchart is shown in fig. 1, that is, a monomer coordinate in a cluster is initialized by using reverse learning; then an IRICP vector searching method is introduced to search for an optimal laser particle control monomer; meanwhile, Laplace chaotic interference is carried out on the optimal laser particle control monomer in the current cluster, so that the optimal monomer of the optimal laser particle in the cluster is obtained.
Specifically, the step of initializing the coordinates of the single body in the cluster by reverse learning comprises:
setting the population scale in the cluster as N, wherein any laser particle control monomer is as follows:
x∈[l,u]；
l is a Laplace space, u is a heat source space, and x is mapped into the two compound spaces for calculation. Its inverse solution x' is:
x′＝l+u-x；
if any candidate laser particle control monomer in the D-dimension vector searching space is P, then:
P＝(x1,x2,…,xd)；
wherein xi∈[ai,bi],i＝1,2,…,d，ai、biIs the digital vector of the controlled object, d is the module value of the vector; then its corresponding inverse solution P' is:
P′＝(x1′,x2′,…,x′d)；
wherein xi′＝li+ui-xi；
And performing reverse learning initialization to obtain the corresponding fitness in the cluster, sequencing the fitness values, and selecting the first N laser particle control monomers corresponding to the fitness values as the initial cluster.
Specifically, the step of searching for the optimal laser particle control monomer group by introducing an IRICP (infrared radiation plasma inductively coupled plasma) vector searching method comprises the following steps:
(1) setting:
initial laser particle control monomer x before reverse learning initialization1，x1∈Rn(ii) a R is a real number;
unit orthogonal direction dnTaking a coordinate direction;
step size
the expansion factor alpha, alpha is more than 1;
reduction factor beta, β ∈ (-1, 0);
the allowable error epsilon is more than 0 and can be determined by a user; order:
i is 1 to n;
in the formula, y1Controlling a monomer for the initialized initial laser particles after reverse learning; n is a positive integer; k is a serial number and takes the value of 1 to n; i and j are laser particle control monomer variables which change along with the setting of a user, wherein j is 1 to n, and i is 1 to n;
(2) laser particle control monomer y after reverse learning initialization with interference in actual judgmentjWhether the corresponding fitness value is smaller than the fitness value corresponding to the initial laser particle control monomer after the reverse learning initialization, namely f (y)j+δjdj)＜f(yj)(δjdjIs an actual interference item), if yes, the laser particle control monomer y after the reverse learning initialization is enabledjThe next laser particle control monomer is equal to the current laser particle control monomer, and the laser particle control monomer yjis enlarged by a factor of a, i.e. yj+1＝yj，δj＝αδj(ii) a Otherwise, the laser particle control monomer y after the reverse learning initialization is enabledjThe next laser particle control monomer is equal to the current laser particle control monomer, and the laser particle control monomer yjby a factor of beta, i.e. yj +1＝yj，δj＝βδj；
(3) Judging a laser particle control monomer variable j, if the laser particle control monomer variable j is smaller than n, namely j is smaller than n, enabling the next laser particle control monomer variable of the laser particle control monomer variable j to be equal to the current laser particle control monomer variable, namely j equals to j +1, and returning to the step (2); otherwise, entering the step (4);
(4) judging whether the fitness value corresponding to the laser particle control single body of the n +1 point after the reverse learning initialization is smaller than the fitness value corresponding to the initial laser particle control single body after the reverse learning initialization, namely f (y)n+1)＜f(y1) If yes, the initial laser particle control monomer after the reverse learning initialization is equal to the laser particle control monomer at the n +1 point after the reverse learning initialization, namely y1＝yn+1J is 1, and return to step (2); if the fitness value corresponding to the laser particle control single body of the n +1 point after the reverse learning initialization is equal to the fitness value corresponding to the initial laser particle control single body after the reverse learning initialization, namely f (y)n +1)＝f(y1) Entering the step (5);
(5) judging whether the fitness value corresponding to the laser particle control monomer of the n +1 point after the reverse learning initialization is smaller than the fitness value corresponding to the laser particle control monomer of the k point after the reverse learning initialization, namely f (y)n+1)＜f(yk) Entering the step (6); otherwise, judging all laser particle control monomer variables j, and if the step length of the point of the laser particle control monomer variable j is not larger than the allowable error, namely | delta | (delta [)jIf | ≦ epsilon, ending, and taking the laser particle control monomer at the k point before the reverse learning initialization as the optimal monomer, or else, taking the initial laser particle control monomer after the reverse learning initialization equal to the laser at the n +1 point after the reverse learning initializationParticle control monomers, i.e. y1＝yn+1And making the laser particle control monomer variable j equal to 1, and returning to the step (2);
(6) the initial laser particle control monomer of k +1 point before the reverse learning initialization is equal to the initial laser particle control monomer of n +1 point after the reverse learning initialization, namely xk+1＝yn+1Judging whether the corresponding fitness value of the initial laser particle control single body of the k +1 point before the reverse learning initialization and the initial laser particle control single body of the k point before the reverse learning initialization is not more than an allowable error, namely | xk+1-xkIf | | is less than or equal to epsilon, taking the initial laser particle control monomer of the k +1 point before the reverse learning initialization as a minimum point and a control object, and finishing the calculation; otherwise, entering the step (7);
(7) order:
in the formula, λiAll along the unit orthogonal direction diAlgebraic sum of step sizes of (a);
defining a set of directions p1,p2,...,pn：
Using Gram-Schmidt orthogonalization method to set the direction { p }jOrthogonalization, let:
then the direction group { pjUnitization, let:
controlling unit orthogonal direction average of a single j point for laser particles; obtaining n new orthogonal vector searching directions according to the formula;
(8) order:
wherein j is 1 to n; and j is 1, then:
y1＝xk+1；
k＝k+1；
and (4) returning to the step (2).
In the above steps, in the case that the initial value and the termination condition are set, the present embodiment may perform the comparison in multiple clusters by performing the loop iteration and the comparison in limited steps, or may perform the comparison in one cluster, thereby obtaining the solution of the optimal monomer, and achieving the requirement of high precision.
Specifically, Laplace chaotic interference is carried out on the optimal laser particle control monomer in the current cluster, the probability of the occurrence of the former superscript controlled phenomenon in the algorithm can be reduced through an interference strategy, and meanwhile cluster diversity can be maintained. The steps proposed in this embodiment include:
for the optimal monomer in the current clusterThe laplace interference strategy is implemented as follows:
in the formula,representing the coordinates of the disturbed monomer, wherein Gaussion (sigma) is a random variable satisfying Laplace distribution; the composite optimal coordinate is updated as follows:
in the formula,controlling the fitness value corresponding to the coordinates of the monomer for the interfered laser particles;controlling monomer for optimal laser particles at time t in current clusterA corresponding fitness value.
By aiming at the current optimal monomerAnd interference operation is carried out, so that the simplex optimal value can be skipped out (if the current composite optimal value is the simplex optimal value), the vector searching efficiency of the algorithm can be effectively improved, and the optimal monomer of the optimal laser particles in the cluster can be obtained.
Specifically, the method of the present invention is verified by selecting the design control problem of the tension-compression nanometer elastic potential energy device.
When the method is used for solving the design control problem of the tension-compression nanometer elastic potential energy device, the constraint is processed firstly. The embodiment selects the tournament selection operator processing constraint condition based on the feasibility rule. The design problem structure of the tension-compression nanometer elastic potential energy device is shown in fig. 2, and the design goal is to minimize the mass of the tension-compression nanometer elastic potential energy device under the constraint conditions of meeting the vibration frequency, the shear stress, the minimum deflection and the like. The design variables are:
diameter d (x) of coil of nanometer elastic potential energy device1) Average diameter D (x) of nano elastic potential energy device ring2) And the number of effective turns P (x)3)。
The target function and the constraint condition of the design control problem of the tension and compression nanometer elastic potential energy device are as follows:
in the formula, x is more than or equal to 0.251≤1.3，0.05≤x2≤2.0，2≤x3≤15。
The design control problem of the tension-compression nanometer elastic potential energy device is solved by using the method, and the parameters are set as follows:
cluster size N is 20, absorption coefficient γ is 1, maximum lead is 0.20, random step size is 0.25, probability of variation pmThe maximum number of cyclic vector seeking times is 1000 as 0.1.
This example compares several representative intelligent control methods, namely, Generic Algorithms (GA), Self-adaptive dependent adaptive approach (SAPA), CPSO, Cooperational differential adaptive (CDE), Mine Blast Algorithm (MBA) adaptive approach-off model (AATM), and Water Cycle Algorithm (WCA), with the eight-cluster intelligent method results shown in FIGS. 3 and 4. In the figure, the MLP represents the data obtained by the method of the present invention. It can be seen that: compared with the SAPA, GA, CPSO and AATM algorithms, the method has the advantages that better results are obtained; compared with the CDE algorithm, the method obtains better optimal results and standard deviation; the present invention achieves better average, worst, and standard deviation results compared to WCA and MBA.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that those skilled in the art can make various improvements and modifications without departing from the principle of the present invention, and these improvements and modifications should also be construed as the protection scope of the present invention.
Claims (3)
1. A laser particle control method is characterized in that a reverse learning is adopted to initialize the monomer coordinates in a cluster; then introducing heat source type feedback control, namely an IRICP (infrared radiation plasma inductively coupled plasma) vector searching method to search an optimal laser particle control monomer group; meanwhile, Laplace chaotic interference is carried out on the optimal laser particle control monomer group in the current cluster, so that the optimal monomer of the optimal laser particle control monomer group in the cluster is obtained;
the method for searching the optimal laser particle control monomer group by introducing an IRICP (infrared inductively coupled plasma) vector searching method comprises the following steps:
(1) setting:
initial laser particle control monomer x before reverse learning initialization1，x1∈Rn；
Unit orthogonal direction dnTaking a coordinate direction;
step size
the expansion factor alpha, alpha is more than 1;
reduction factor beta, β ∈ (-1, 0);
the allowable error epsilon is more than 0; order:
y1＝x1,k＝1,j＝1,i is 1 to n;
in the formula, y1Controlling a monomer for the initialized initial laser particles after reverse learning; k is a serial number, k is 1 to n; i and j are laser particle control monomer variables, j is 1 to n, and i is 1 to n;
(2) laser particle control monomer y after reverse learning initialization with interference in actual judgmentjWhether the corresponding fitness value is smaller than the fitness value corresponding to the initial laser particle control monomer after the reverse learning initialization or not is judged, if so, the laser particle control monomer y after the reverse learning initialization is enabledjThe next laser particle control monomer is equal to the current laser particle control monomer, and the laser particle control monomer yjstep length of (2) is enlarged by alpha times, otherwise, the laser particle control monomer y after the reverse learning initialization is enabledjThe next laser particle control monomer is equal to the current laser particle control monomer, and the laser particle control monomer yjthe step size of (2) is reduced by beta times;
(3) judging a laser particle control monomer variable j, if the laser particle control monomer variable j is smaller than n, enabling the next laser particle control monomer variable of the laser particle control monomer variable j to be equal to the current laser particle control monomer variable, and returning to the step (2); otherwise, entering the step (4);
(4) judging whether the fitness value corresponding to the laser particle control monomer of the n +1 point after the reverse learning initialization is smaller than the fitness value corresponding to the initial laser particle control monomer after the reverse learning initialization, if so, making the initial laser particle control monomer after the reverse learning initialization equal to the laser particle control monomer of the n +1 point after the reverse learning initialization, and returning to the step (2); if the fitness value corresponding to the laser particle control single body of the n +1 point after the reverse learning initialization is equal to the fitness value corresponding to the initial laser particle control single body after the reverse learning initialization, entering the step (5);
(5) judging whether the fitness value corresponding to the laser particle control monomer of the n +1 point after the reverse learning initialization is smaller than the fitness value corresponding to the laser particle control monomer of the k point after the reverse learning initialization, and if so, entering the step (6); otherwise, judging all laser particle control monomer variables j, if the step length of the point of the laser particle control monomer variable j is not larger than the allowable error, ending, taking the laser particle control monomer at the point k before the reverse learning initialization as the optimal monomer, otherwise, taking the initial laser particle control monomer after the reverse learning initialization to be equal to the laser particle control monomer at the point n +1 after the reverse learning initialization, making the laser particle control monomer variable j equal to 1, and returning to the step (2);
(6) making the initial laser particle control monomer of the k +1 point before the reverse learning initialization equal to the initial laser particle control monomer of the n +1 point after the reverse learning initialization, judging that if the corresponding fitness value of the initial laser particle control monomer of the k +1 point before the reverse learning initialization and the corresponding fitness value of the initial laser particle control monomer of the k point before the reverse learning initialization are not more than an allowable error, taking the initial laser particle control monomer of the k +1 point before the reverse learning initialization as a minimum point and a control object, and finishing the calculation; otherwise, entering the step (7);
(7) order:
in the formula, λiAll along the unit orthogonal direction diAlgebraic sum of step sizes of (a);
defining a set of directions p1,p2,...,pn：
Using Gram-Schmidt orthogonalization method to set the direction { p }jOrthogonalizing, let the direction qjThe determination is as follows:
then the direction group { pjUnitization, let:
controlling unit orthogonal direction average of a single j point for laser particles; obtaining n new orthogonal vector searching directions according to the formula;
(8) order:
wherein j is 1 to n; and j is 1, then:
y1＝xk+1；
k＝k+1；
and (4) returning to the step (2).
2. The laser particle control method of claim 1, wherein the initializing coordinates of the individual in the cluster using inverse learning comprises:
setting the population scale in the cluster as N, wherein any laser particle control monomer is as follows:
x∈[l,u]；
its inverse solution x' is:
x′＝l+u-x；
if any candidate laser particle control monomer in the D-dimension vector searching space is P, then:
P＝(x1,x2,…,xd)；
wherein xi∈[ai,bi]I is 1,2, …, d; then its corresponding inverse solution P' is:
P′＝(x′1,x′2,…,x′d)；
wherein x'i＝li+ui-xi；
And performing reverse learning initialization to obtain the corresponding fitness in the cluster, sequencing the fitness values, and selecting the first N laser particle control monomers corresponding to the fitness values as the initial cluster.
3. The laser particle control method of claim 1, wherein the step of performing laplacian chaotic disturbance on the optimal laser particle control single in the current cluster comprises:
control monomer for optimal laser particles at time t in current clusterThe laplace interference strategy is implemented as follows:
in the formula,representing the coordinates of the laser particle control monomer after the interference, wherein Gaussion (sigma) is a random variable satisfying Laplace distribution; the composite optimal coordinate is updated as follows:
in the formula,controlling the fitness value corresponding to the coordinates of the monomer for the interfered laser particles;controlling monomer for optimal laser particles at time t in current clusterA corresponding fitness value;
by controlling the monomer for the current optimum laser particleAnd carrying out interference operation, thereby obtaining the optimal monomer of the optimal laser particle control monomer group in the cluster.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710752460.3A CN107728469B (en) | 2017-08-28 | 2017-08-28 | A kind of laser particle control method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710752460.3A CN107728469B (en) | 2017-08-28 | 2017-08-28 | A kind of laser particle control method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107728469A CN107728469A (en) | 2018-02-23 |
| CN107728469B true CN107728469B (en) | 2018-08-24 |
Family
ID=61205464
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710752460.3A Active CN107728469B (en) | 2017-08-28 | 2017-08-28 | A kind of laser particle control method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107728469B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110095981A (en) * | 2019-04-02 | 2019-08-06 | 南京交通职业技术学院 | A kind of setting method, device and the electronic equipment of automatic disturbance rejection controller parameter |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102968665A (en) * | 2012-12-05 | 2013-03-13 | 苏州大学 | Forward kinematics solving method of parallel robot |
| CN106447024A (en) * | 2016-08-31 | 2017-02-22 | 上海电机学院 | Particle swarm improved algorithm based on chaotic backward learning |
| CN106501801A (en) * | 2016-09-28 | 2017-03-15 | 哈尔滨工程大学 | A kind of bistatic MIMO radar tracking based on chaos Symbiotic evolution on multiple populations |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6909102B1 (en) * | 2004-01-21 | 2005-06-21 | Varian Semiconductor Equipment Associates, Inc. | Ion implanter system, method and program product including particle detection |
| JP5374731B2 (en) * | 2008-11-26 | 2013-12-25 | 独立行政法人日本原子力研究開発機構 | Laser-driven particle beam irradiation apparatus and method of operating laser-driven particle beam irradiation apparatus |
| CN103611928B (en) * | 2013-12-05 | 2016-01-20 | 东北大学 | A kind of silver and silver composite nano particles and laser controlling synthetic method thereof |
| CN105654476B (en) * | 2015-12-25 | 2019-03-08 | 江南大学 | Binocular calibration method based on Chaos particle swarm optimization algorithm |
| CN105842142A (en) * | 2016-05-18 | 2016-08-10 | 深圳市青核桃科技有限公司 | Method for calibrating laser particle counter by using single standard particle |
| CN106209706B (en) * | 2016-06-28 | 2019-04-05 | 电子科技大学 | The detection method of weak harmonic signal under a kind of Chaotic Background |
| CN106355192B (en) * | 2016-08-16 | 2019-12-31 | 温州大学 | Support vector machine method based on chaos grey wolf optimization |
-
2017
- 2017-08-28 CN CN201710752460.3A patent/CN107728469B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102968665A (en) * | 2012-12-05 | 2013-03-13 | 苏州大学 | Forward kinematics solving method of parallel robot |
| CN106447024A (en) * | 2016-08-31 | 2017-02-22 | 上海电机学院 | Particle swarm improved algorithm based on chaotic backward learning |
| CN106501801A (en) * | 2016-09-28 | 2017-03-15 | 哈尔滨工程大学 | A kind of bistatic MIMO radar tracking based on chaos Symbiotic evolution on multiple populations |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107728469A (en) | 2018-02-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | A policy search method for temporal logic specified reinforcement learning tasks | |
| Li et al. | A survey on firefly algorithms | |
| Li et al. | Hybrid optimization algorithm based on chaos, cloud and particle swarm optimization algorithm | |
| CN110530373B (en) | Robot path planning method, controller and system | |
| Zhai et al. | Instance selection for time series classification based on immune binary particle swarm optimization | |
| CN106447022A (en) | Multi-objective particle swarm optimization based on globally optimal solution selected according to regions and individual optimal solution selected according to proximity | |
| Yin et al. | A novel SVM parameter tuning method based on advanced whale optimization algorithm | |
| Ren | Optimal control | |
| CN115373400A (en) | Robot path planning method and system based on dynamic update mechanism ant colony algorithm | |
| Peng et al. | A new fuzzy adaptive simulated annealing genetic algorithm and its convergence analysis and convergence rate estimation | |
| Aoun et al. | Particle swarm optimisation with population size and acceleration coefficients adaptation using hidden Markov model state classification | |
| Zhao et al. | Optimizing widths with PSO for center selection of Gaussian radial basis function networks | |
| CN110533151A (en) | A kind of firefly optimization algorithm based on the law of universal gravitation | |
| CN107728469B (en) | A kind of laser particle control method | |
| Ding et al. | Gradient-based meta-learning using uncertainty to weigh loss for few-shot learning | |
| CN117420832A (en) | Robot path planning method based on improved GTO | |
| Chen et al. | Bare-bones imperialist competitive algorithm for a compensatory neural fuzzy controller | |
| Lin et al. | A recurrent neural fuzzy controller based on self‐organizing improved particle swarm optimization for a magnetic levitation system | |
| Wang et al. | Improved grey wolf optimizer for multiple unmanned aerial vehicles task allocation | |
| Li et al. | MCMC: Multi-Constrained Model Compression via One-Stage Envelope Reinforcement Learning | |
| Yang et al. | Hybrid Taguchi-based particle swarm optimization for flowshop scheduling problem | |
| CN111860755A (en) | Improved particle swarm algorithm based on regression of support vector machine | |
| Liang et al. | Improved grey wolf optimizer and their applications | |
| Akbari et al. | Combination of particle swarm optimization and stochastic local search for multimodal function optimization | |
| Feng et al. | An improved hybrid encoding firefly algorithm for randomized time-varying knapsack problems |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20180717 Address after: 361021 one of the three floors of No. 115 workshop of Tianan Road, Jimei District, Xiamen, Fujian. Applicant after: Three technologies (Xiamen) Electronic Technology Co. Ltd. Address before: 361021 one of the three floors of No. 115 workshop of Tianan Road, Jimei District, Xiamen, Fujian. Applicant before: Three technologies (Xiamen) Electronic Technology Co. Ltd. Applicant before: Harvest Technology (Beijing) Co., Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |